Before measuring the dimensions of the four faces of the main bearing cover of the engine, it is necessary to identify the four faces of the bearing cover first, and then distinguish the four faces of the bearing cover. Based on the feature extraction capabilities and recognition and classification capabilities of the convolutional neural network, a convolutional neural network-based recognition algorithm for the engine main bearing cover surface is proposed. This algorithm removes the traditional complex preprocessing operations and extracts four surfaces of the bearing cover. To identify the four faces of the bearing cap. Experimental results show that the algorithm proposed in this paper can correctly identify the four faces of the main bearing cover of the engine, with an average recognition accuracy of 100% and an average recognition time of 3.80s. It has the characteristics of high recognition accuracy, short recognition time, and strong anti-interference ability.